An Engineered IgG-VHH Bispecific Antibody against SARS-CoV-2 and Its Variants.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) neutralizing antibodies are shown to be effective therapeutics for providing coronavirus disease 2019 (COVID-19) protection. However, recurrent variants arise and facilitate significant escape from current antibody therapeutics. Bispecific antibodies (bsAbs) represent a unique platform to increase antibody breadth and to reduce neutralization escape. Herein, a novel immunoglobulin G-variable domains of heavy-chain-only antibody (IgG-VHH) format bsAb derived from a potent human antibody R15-F7 and a humanized nanobody P14-F8-35 are rationally engineered. The resulting bsAb SYZJ001 efficiently neutralizes wild-type SARS-CoV-2 as well as the alpha, beta, gamma, and delta variants, with superior efficacy to its parental antibodies. Cryo-electron microscopy structural analysis reveals that R15-F7 and P14-F8-35 bind to nonoverlapping epitopes within the RBD and sterically hindered ACE2 receptor binding. Most importantly, SYZJ001 shows potent prophylactic and therapeutic efficacy against SARS-CoV-2 in three established mouse models. Collectively, the current results demonstrate that the novel bsAb format is feasible and effective, suggesting great potential as an inspiring antiviral strategy.

Autoimmune Skin Disease Exacerbations Following COVID-19 Vaccination.

Background: Vaccination against COVID-19 reduces the risk of severe COVID-19 disease and death. However, few studies have examined the safety of the COVID-19 vaccine in patients with autoimmune skin disease. Objectives: We sought to determine the incidence of disease exacerbation in this population following COVID-19 vaccination as well as the associated factors. Methods: We performed a chart review of all patients seen in the autoimmune skin disease clinic of the principal investigator during the study period. All patients included for analysis were systematically and prospectively asked about COVID-19 vaccination status, manufacturers, vaccine dates, autoimmune symptoms after the vaccine, and timing of symptom onset using a standardized template as part of their visit. Demographics and autoimmune disease diagnosis were also collected. Analysis used Chi-square and Fisher's exact tests. Results: 402 subjects were included for analysis. 85.6% of patients were fully vaccinated, with 12.9% unvaccinated and 1.5% partially vaccinated. 14.8% of fully vaccinated patients reported worsening autoimmune signs and symptoms after the vaccine. Fully vaccinated dermatomyositis patients were more likely to report worsening autoimmune signs and symptoms after the vaccine (22.7%) than fully vaccinated lupus erythematosus patients (8.6%) (p=0.009). Patients fully vaccinated with the Moderna vaccine trended towards an increased likelihood of reporting worsening autoimmune signs and symptoms after the vaccine (19.1%) than those with the Pfizer-BioNTech vaccine (12.0%) (p=0.076). Of the patients who had autoimmune symptoms after vaccination, 20% had symptoms after the 1st dose, 82% after the 2nd dose, and 4% after the 3rd dose with median onset (95% confidence interval) of 7 (2,14), 14 (14,21), and 18 (7,28) days later, respectively. Conclusions: More fully vaccinated dermatomyositis patients had exacerbation of autoimmune signs and symptoms after the vaccine than fully vaccinated lupus erythematosus patients. However, given the risks of COVID-19, clinicians should still promote vaccination in most patients with autoimmune skin disease.

BA.2.12.1, BA.4 and BA.5 escape antibodies elicited by Omicron infection.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron sublineages BA.2.12.1, BA.4 and BA.5 exhibit higher transmissibility than the BA.2 lineage1. The receptor binding and immune-evasion capability of these recently emerged variants require immediate investigation. Here, coupled with structural comparisons of the spike proteins, we show that BA.2.12.1, BA.4 and BA.5 (BA.4 and BA.5 are hereafter referred collectively to as BA.4/BA.5) exhibit similar binding affinities to BA.2 for the angiotensin-converting enzyme 2 (ACE2) receptor. Of note, BA.2.12.1 and BA.4/BA.5 display increased evasion of neutralizing antibodies compared with BA.2 against plasma from triple-vaccinated individuals or from individuals who developed a BA.1 infection after vaccination. To delineate the underlying antibody-evasion mechanism, we determined the escape mutation profiles2, epitope distribution3 and Omicron-neutralization efficiency of 1,640 neutralizing antibodies directed against the receptor-binding domain of the viral spike protein, including 614 antibodies isolated from people who had recovered from BA.1 infection. BA.1 infection after vaccination predominantly recalls humoral immune memory directed against ancestral (hereafter referred to as wild-type (WT)) SARS-CoV-2 spike protein. The resulting elicited antibodies could neutralize both WT SARS-CoV-2 and BA.1 and are enriched on epitopes on spike that do not bind ACE2. However, most of these cross-reactive neutralizing antibodies are evaded by spike mutants L452Q, L452R and F486V. BA.1 infection can also induce new clones of BA.1-specific antibodies that potently neutralize BA.1. Nevertheless, these neutralizing antibodies are largely evaded by BA.2 and BA.4/BA.5 owing to D405N and F486V mutations, and react weakly to pre-Omicron variants, exhibiting narrow neutralization breadths. The therapeutic neutralizing antibodies bebtelovimab4 and cilgavimab5 can effectively neutralize BA.2.12.1 and BA.4/BA.5, whereas the S371F, D405N and R408S mutations undermine most broadly sarbecovirus-neutralizing antibodies. Together, our results indicate that Omicron may evolve mutations to evade the humoral immunity elicited by BA.1 infection, suggesting that BA.1-derived vaccine boosters may not achieve broad-spectrum protection against new Omicron variants.

Autoimmune Skin Disease Exacerbations Following COVID-19 Vaccination

Background Vaccination against COVID-19 reduces the risk of severe COVID-19 disease and death. However, few studies have examined the safety of the COVID-19 vaccine in patients with autoimmune skin disease. Objectives We sought to determine the incidence of disease exacerbation in this population following COVID-19 vaccination as well as the associated factors. Methods We performed a chart review of all patients seen in the autoimmune skin disease clinic of the principal investigator during the study period. All patients included for analysis were systematically and prospectively asked about COVID-19 vaccination status, manufacturers, vaccine dates, autoimmune symptoms after the vaccine, and timing of symptom onset using a standardized template as part of their visit. Demographics and autoimmune disease diagnosis were also collected. Analysis used Chi-square and Fisher’s exact tests. Results 402 subjects were included for analysis. 85.6% of patients were fully vaccinated, with 12.9% unvaccinated and 1.5% partially vaccinated. 14.8% of fully vaccinated patients reported worsening autoimmune signs and symptoms after the vaccine. Fully vaccinated dermatomyositis patients were more likely to report worsening autoimmune signs and symptoms after the vaccine (22.7%) than fully vaccinated lupus erythematosus patients (8.6%) (p=0.009). Patients fully vaccinated with the Moderna vaccine trended towards an increased likelihood of reporting worsening autoimmune signs and symptoms after the vaccine (19.1%) than those with the Pfizer-BioNTech vaccine (12.0%) (p=0.076). Of the patients who had autoimmune symptoms after vaccination, 20% had symptoms after the 1st dose, 82% after the 2nd dose, and 4% after the 3rd dose with median onset (95% confidence interval) of 7 (2,14), 14 (14,21), and 18 (7,28) days later, respectively. Conclusions More fully vaccinated dermatomyositis patients had exacerbation of autoimmune signs and symptoms after the vaccine than fully vaccinated lupus erythematosus patients. However, given the risks of COVID-19, clinicians should still promote vaccination in most patients with autoimmune skin disease.

Transcriptional regulation and small compound targeting of ACE2 in lung epithelial cells.

Angiotensin-converting enzyme 2 (ACE2) is the receptor of COVID-19 pathogen SARS-CoV-2, but the transcription factors (TFs) that regulate the expression of the gene encoding ACE2 (ACE2) have not been systematically dissected. In this study we evaluated TFs that control ACE2 expression, and screened for small molecule compounds that could modulate ACE2 expression to block SARS-CoV-2 from entry into lung epithelial cells. By searching the online datasets we found that 24 TFs might be ACE2 regulators with signal transducer and activator of transcription 3 (Stat3) as the most significant one. In human normal lung tissues, the expression of ACE2 was positively correlated with phosphorylated Stat3 (p-Stat3). We demonstrated that Stat3 bound ACE2 promoter, and controlled its expression in 16HBE cells stimulated with interleukin 6 (IL-6). To screen for medicinal compounds that could modulate ACE2 expression, we conducted luciferase assay using HLF cells transfected with ACE2 promoter-luciferase constructs. Among the 64 compounds tested, 6-O-angeloylplenolin (6-OAP), a sesquiterpene lactone in Chinese medicinal herb Centipeda minima (CM), represented the most potent ACE2 repressor. 6-OAP (2.5 µM) inhibited the interaction between Stat3 protein and ACE2 promoter, thus suppressed ACE2 transcription. 6-OAP (1.25-5 µM) and its parental medicinal herb CM (0.125%-0.5%) dose-dependently downregulated ACE2 in 16HBE and Beas-2B cells; similar results were observed in the lung tissues of mice following administration of 6-OAP or CM for one month. In addition, 6-OAP/CM dose-dependently reduced IL-6 production and downregulated chemokines including CXCL13 and CX3CL1 in 16HBE cells. Moreover, we found that 6-OAP/CM inhibited the entry of SARS-CoV-2 S protein pseudovirus into target cells. These results suggest that 6-OAP/CM are ACE2 inhibitors that may potentially protect lung epithelial cells from SARS-CoV-2 infection.

The atlas of ACE2 expression in fetal and adult human hearts reveals the potential mechanism of heart-injured patients infected with SARS-CoV-2.

Numerous studies have shown that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can infect host cells through binding to angiotensin I converting enzyme 2 (ACE2) expressing in various tissues and organs. In this study, we deeply analyzed the single-cell expression profiles of ACE2 in fetal and adult human hearts to explore the potential mechanism of SARS-CoV-2 harming the heart. The molecular docking software was used to simulate the binding of SARS-CoV-2 and its variant spike protein with ACE2. The genes closely related to ACE2 in renin-angiotensin system (RAS) were identified by constructing a protein-protein interaction network. Through the analysis of single-cell transcription profiles at different stages of human embryos, we found that the expression level of ACE2 in ventricular myocytes was increased with embryonic development. The results of single-cell sequencing analysis showed that the expression of ACE2 in ventricular myocytes was upregulated in heart failure induced by dilated cardiomyopathy compared with normal hearts. The upregulation of ACE2 increases the risk of infection with SARS-CoV-2 in fetal and adult human hearts. We also further confirmed the expression of ACE2 and ACE2-related genes in normal and SARS-CoV-2-infected human pluripotent stem cell-derived cardiomyocytes. In addition, the pathway analysis revealed that ACE2 may regulate the differently expressed genes in heart failure through calcium signaling pathway and Wnt signaling pathway.

Structural and functional characterizations of infectivity and immune evasion of SARS-CoV-2 Omicron.

The SARS-CoV-2 Omicron variant with increased fitness is spreading rapidly worldwide. Analysis of cryo-EM structures of the spike (S) from Omicron reveals amino acid substitutions forging interactions that stably maintain an active conformation for receptor recognition. The relatively more compact domain organization confers improved stability and enhances attachment but compromises the efficiency of the viral fusion step. Alterations in local conformation, charge, and hydrophobic microenvironments underpin the modulation of the epitopes such that they are not recognized by most NTD- and RBD-antibodies, facilitating viral immune escape. Structure of the Omicron S bound with human ACE2, together with the analysis of sequence conservation in ACE2 binding region of 25 sarbecovirus members, as well as heatmaps of the immunogenic sites and their corresponding mutational frequencies, sheds light on conserved and structurally restrained regions that can be used for the development of broad-spectrum vaccines and therapeutics.

Tubular epithelial cell-to-macrophage communication forms a negative feedback loop via extracellular vesicle transfer to promote renal inflammation and apoptosis in diabetic nephropathy.

Background: Macrophage infiltration around lipotoxic tubular epithelial cells (TECs) is a hallmark of diabetic nephropathy (DN). However, how these two types of cells communicate remains obscure. We previously demonstrated that LRG1 was elevated in the process of kidney injury. Here, we demonstrated that macrophage-derived, LRG1-enriched extracellular vesicles (EVs) exacerbated DN. Methods: We induced an experimental T2DM mouse model with a HFD diet for four months. Renal primary epithelial cells and macrophage-derived EVs were isolated from T2D mice by differential ultracentrifugation. To investigate whether lipotoxic TEC-derived EV (EVe) activate macrophages, mouse bone marrow-derived macrophages (BMDMs) were incubated with EVe. To investigate whether activated macrophage-derived EVs (EVm) induce lipotoxic TEC apoptosis, EVm were cocultured with primary renal tubular epithelial cells. Subsequently, we evaluated the effect of LRG1 in EVe by investigating the apoptosis mechanism. Results: We demonstrated that incubation of primary TECs of DN or HK-2 mTECs with lysophosphatidyl choline (LPC) increased the release of EVe. Interestingly, TEC-derived EVe activated an inflammatory phenotype in macrophages and induced the release of macrophage-derived EVm. Furthermore, EVm could induce apoptosis in TECs injured by LPC. Importantly, we found that leucine-rich α-2-glycoprotein 1 (LRG1)-enriched EVe activated macrophages via a TGFßR1-dependent process and that tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-enriched EVm induced apoptosis in injured TECs via a death receptor 5 (DR5)-dependent process. Conclusion: Our findings indicated a novel cell communication mechanism between tubular epithelial cells and macrophages in DN, which could be a potential therapeutic target.

Characterization and structural basis of a lethal mouse-adapted SARS-CoV-2.

There is an urgent need for animal models to study SARS-CoV-2 pathogenicity. Here, we generate and characterize a novel mouse-adapted SARS-CoV-2 strain, MASCp36, that causes severe respiratory symptoms, and mortality. Our model exhibits age- and gender-related mortality akin to severe COVID-19. Deep sequencing identified three amino acid substitutions, N501Y, Q493H, and K417N, at the receptor binding domain (RBD) of MASCp36, during in vivo passaging. All three RBD mutations significantly enhance binding affinity to its endogenous receptor, ACE2. Cryo-electron microscopy analysis of human ACE2 (hACE2), or mouse ACE2 (mACE2), in complex with the RBD of MASCp36, at 3.1 to 3.7 Å resolution, reveals the molecular basis for the receptor-binding switch. N501Y and Q493H enhance the binding affinity to hACE2, whereas triple mutations at N501Y/Q493H/K417N decrease affinity and reduce infectivity of MASCp36. Our study provides a platform for studying SARS-CoV-2 pathogenesis, and unveils the molecular mechanism for its rapid adaptation and evolution.

High-Resolution CT Features of the COVID-19 Infection During Short-Term Follow-Up

Objective: To estimate the short-term follow-up chest HRCT manifestation in patients with COVID-19 who have left hospitals. Methods: Retrospective analysis was executed on the chest HRCT manifestation of 49 cases with RT-PCR confirmed COVID-19 pneumonia patients diagnosed in Beijing You’An hospital affiliated to capital medical university between 20 January 2020 to April 10, 2020. The study including 21 males and 28 females, and the median age was 59.5 years age ranged from 21 years to 80 years. The evolution characteristics of chest HRCT in short-term follow-up after discharge were analyzed. Results: Forty-nine patients were divided into two groups according to Diagnosis and Treatment Protocol of 2019 New Coronavirus Pneumonia (trial version 7) of National Health Commission of China. Severe group included 23 cases (46.9%) and 26 cases of ordinary (53.1%). The HRCT findings of patients in the 2 groups were analyzed on median follow-up time was 30 day (range, 14 to 46 days) after discharge. The follow-up chest HRCT were normal in 19 of 49 (38.8%) patients, abnormal HRCT findings were detected in 30 of 49 (61.2%) patients. The most frequent abnormality found on follow-up HRCT in COVID-19 patients who recovered from pneumonia was GGO which was seen in 26 of 49 (53.1%) patients, reticular pattern was in 9 of 49 (18.4%), traction bronchiectasis was in 3 of 49 (6.1%), small nodular consolidation was in 6 of 49 (12.2%), other findings including curved line in the subpleural area and irregular linear opacities were in 15 of 49 (30.6%). In the study, 3 of 49 (6.1%) patients were diagnosed with pulmonary fibrosis according to HRCT findings. Three cases were severe patients, aged from 65 to 80 years. Pulmonary fibrosis occurred within 30–50 days of onset. Conclusion: Pulmonary fibrosis may develop in 6.1% of COVID-19 patients, which was lower than that of SARS and MERS, suggesting that the lung damage caused by SARS-COV-2 in the recovered patients was lower than that of SARS and MERS.

Rational development of a human antibody cocktail that deploys multiple functions to confer Pan-SARS-CoVs protection.

Structural principles underlying the composition and synergistic mechanisms of protective monoclonal antibody cocktails are poorly defined. Here, we exploited antibody cooperativity to develop a therapeutic antibody cocktail against SARS-CoV-2. On the basis of our previously identified humanized cross-neutralizing antibody H014, we systematically analyzed a fully human naive antibody library and rationally identified a potent neutralizing antibody partner, P17, which confers effective protection in animal model. Cryo-EM studies dissected the nature of the P17 epitope, which is SARS-CoV-2 specific and distinctly different from that of H014. High-resolution structure of the SARS-CoV-2 spike in complex with H014 and P17, together with functional investigations revealed that in a two-antibody cocktail, synergistic neutralization was achieved by S1 shielding and conformational locking, thereby blocking receptor attachment and viral membrane fusion, conferring high potency as well as robustness against viral mutation escape. Furthermore, cluster analysis identified a hypothetical 3rd antibody partner for further reinforcing the cocktail as pan-SARS-CoVs therapeutics.

A proof of concept for neutralizing antibody-guided vaccine design against SARS-CoV-2.

Mutations and transient conformational movements of the receptor binding domain (RBD) that make neutralizing epitopes momentarily unavailable present immune escape routes for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). To mitigate viral escape, we developed a cocktail of neutralizing antibodies (NAbs) targeting epitopes located on different domains of spike (S) protein. Screening of a library of monoclonal antibodies generated from peripheral blood mononuclear cells of COVID-19 convalescent patients yielded potent NAbs, targeting the N-terminal domain (NTD) and RBD domain of S, effective at nM concentrations. Remarkably, a combination of RBD-targeting NAbs and NTD-binding NAbs, FC05, enhanced the neutralization potency in cell-based assays and an animal model. Results of competitive surface plasmon resonance assays and cryo-electron microscopy (cryo-EM) structures of antigen-binding fragments bound to S unveil determinants of immunogenicity. Combinations of immunogens, identified in the NTD and RBD of S, when immunized in rabbits and macaques, elicited potent protective immune responses against SARS-CoV-2. More importantly, two immunizations of this combination of NTD and RBD immunogens provided complete protection in macaques against a SARS-CoV-2 challenge, without observable antibody-dependent enhancement of infection. These results provide a proof of concept for neutralization-based immunogen design targeting SARS-CoV-2 NTD and RBD.

Unknown Disease Outbreaks Detection: A Pilot Study on Feature-Based Knowledge Representation and Reasoning Model.

Background: The outbreak of COVID-19 in 2019 has rapidly swept the world, causing irreparable loss to human beings. The pandemic has shown that there is still a delay in the early response to disease outbreaks and needs a method for unknown disease outbreak detection. The study's objective is to establish a new medical knowledge representation and reasoning model, and use the model to explore the feasibility of unknown disease outbreak detection. Methods: The study defined abnormal values with diagnostic significances from clinical data as the Features, and defined the Features as the antecedents of inference rules to match with knowledge bases, achieved in detecting known or emerging infectious disease outbreaks. Meanwhile, the study built a syndromic surveillance base to capture the target cases' Features to improve the reliability and fault-tolerant ability of the system. Results: The study combined the method with Severe Acute Respiratory Syndrome (SARS), Middle East Respiratory Syndrome (MERS), and early COVID-19 outbreaks as empirical studies. The results showed that with suitable surveillance guidelines, the method proposed in this study was capable to detect outbreaks of SARS, MERS, and early COVID-19 pandemics. The quick matching accuracies of confirmed infection cases were 89.1, 26.3-98%, and 82%, and the syndromic surveillance base would capture the Features of the remaining cases to ensure the overall detection accuracies. Based on the early COVID-19 data in Wuhan, this study estimated that the median time of the early COVID-19 cases from illness onset to local authorities' responses could be reduced to 7.0-10.0 days. Conclusions: This study offers a new solution to transfer traditional medical knowledge into structured data and form diagnosis rules, enables the representation of doctors' logistic thinking and the knowledge transmission among different users. The results of empirical studies demonstrate that by constantly inputting medical knowledge into the system, the proposed method will be capable to detect unknown diseases from existing ones and perform an early response to the initial outbreaks.

Periphery-aware COVID-19 diagnosis with contrastive representation enhancement.

Computer-aided diagnosis has been extensively investigated for more rapid and accurate screening during the outbreak of COVID-19 epidemic. However, the challenge remains to distinguish COVID-19 in the complex scenario of multi-type pneumonia classification and improve the overall diagnostic performance. In this paper, we propose a novel periphery-aware COVID-19 diagnosis approach with contrastive representation enhancement to identify COVID-19 from influenza-A (H1N1) viral pneumonia, community acquired pneumonia (CAP), and healthy subjects using chest CT images. Our key contributions include: 1) an unsupervised Periphery-aware Spatial Prediction (PSP) task which is designed to introduce important spatial patterns into deep networks; 2) an adaptive Contrastive Representation Enhancement (CRE) mechanism which can effectively capture the intra-class similarity and inter-class difference of various types of pneumonia. We integrate PSP and CRE to obtain the representations which are highly discriminative in COVID-19 screening. We evaluate our approach comprehensively on our constructed large-scale dataset and two public datasets. Extensive experiments on both volume-level and slice-level CT images demonstrate the effectiveness of our proposed approach with PSP and CRE for COVID-19 diagnosis.

Impact of a pandemic on surgical neuro-oncology-maintaining functionality in the early phase of crisis.

BACKGROUND: The Coronavirus disease 2019 (COVID-19) pandemic has extraordinarily impacted global healthcare. Neuro-oncological surgery units have peculiar features that make them highly relevant in the strategic reaction to the pandemic. In this Chinese Society of Neuro-Oncology (CSNO) initiated survey, we appraise the changes implemented in neuro-oncological surgery hospitals across different Asian countries and provide expert recommendations for responses at different stages of the pandemic. METHODS: We performed a 42-question survey of the early experience of neuro-oncological surgery practice in hospitals across different Asian countries on April 1, 2020, with responses closed on April 18, 2020. RESULTS: 144 hospitals completed the questionnaire. Most were in WHO post-peak phase of the pandemic and reported a median reduction in neuro-oncological surgery volume of 25-50%. Most (67.4%) resumed elective surgery in only COVID-19 negative patients;11.1% performed only emergency cases irrespective of COVID-19 status;2.1% suspended all surgical activity. Ninety-one (63.2%) relocated personnel from neurosurgery to other departments. Fifty-two (36.1%) hospitals suspended post-operative adjuvant therapy and 94 (65.2%) instituted different measures to administer post-operative adjuvant therapy. Majority (59.0%) of the hospitals suspended research activity. Most (70%) respondents anticipate that current neurosurgery restrictions will continue to remain for > 1 month. CONCLUSIONS: Majority of the respondents to our survey reported reduced neuro-oncological surgery activity, policy modification, personnel reallocation, and curtailment of educational/research activities in response to the COVID-19 pandemic. The persistent widespread interruption of surgical neuro-oncology in even post-peak phases of the pandemic raises serious concerns about the long-term impact of the pandemic on neuro-oncological patients and highlights the essence of timely measures for pandemic preparedness, patient triage, and workforce protection.

Down-Regulation of Colonic ACE2 Expression in Patients With Inflammatory Bowel Disease Responding to Anti-TNF Therapy: Implications for COVID-19.

Background and Aims: Angiotensin-converting enzyme II (ACE2) is the key molecule for understanding the pathophysiology of COVID-19. The risk of COVID-19 and impact of immunosuppressive treatment on disease course in patients with inflammatory bowel disease (IBD) remain controversial. We aimed to determine the change of intestinal ACE2 expression before and after biologics treatment including anti-tumor necrosis factor α (anti-TNFα), anti-integrin, and anti-interleukin (IL)12/23 in IBD patients. Methods: We analyzed the ACE2 expression through the public database of paired intestinal biopsies from IBD patients before and after biologic therapy. Change of ACE2 RNA and protein expression were validated in two independent cohorts (Birmingham cohort and Guangzhou cohort). The correlation between ACE2 expression and disease activity was also analyzed. Results: Mining information from the GEO database showed that compared with healthy control, intestinal ACE2 expression was downregulated in ileum of CD patients, while upregulated in colon of both CD and UC patients. Colonic ACE2 RNA expression was decreased significantly in patients responding to anti-TNFα but not anti-integrin and anti-IL12/23, which was validated in the Birmingham cohort. Using the Guangzhou cohort including 53 patients matched by pre- and post-anti-TNFα therapy, colonic ACE2 protein expression was significantly downregulated after anti-TNFα treatment in responders (P < 0.001) rather than non-responders. Colonic ACE2 expression was significantly higher in patients with severe histologically active disease compared with those with moderate (P < 0.0001) and mild (P = 0.0002) histologically active disease. Conclusion: Intestinal inflammation influences the expression of intestinal ACE2 in IBD patients, with different alterations in the ileum and colon. Colonic ACE2 expression was downregulated after anti-TNFα therapy in IBD patients responding to treatment. This might provide new clues regarding the risk of SARS-CoV-2 infection and the potential benefit of sustaining anti-TNFα treatment in patients with IBD.

Deep learning for detecting corona virus disease 2019 (COVID-19) on high-resolution computed tomography: a pilot study.

BACKGROUND: To evaluate the diagnostic efficacy of Densely Connected Convolutional Networks (DenseNet) for detection of COVID-19 features on high resolution computed tomography (HRCT). METHODS: The Ethic Committee of our institution approved the protocol of this study and waived the requirement for patient informed consent. Two hundreds and ninety-five patients were enrolled in this study (healthy person: 149; COVID-19 patients: 146), which were divided into three separate non-overlapping cohorts (training set, n=135, healthy person, n=69, patients, n=66; validation set, n=20, healthy person, n=10, patients, n=10; test set, n=140, healthy person, n=70, patients, n=70). The DenseNet was trained and tested to classify the images as having manifestation of COVID-19 or as healthy. A radiologist also blindly evaluated all the test images and rechecked the misdiagnosed cases by DenseNet. Receiver operating characteristic curves (ROC) and areas under the curve (AUCs) were used to assess the model performance. The sensitivity, specificity and accuracy of DenseNet model and radiologist were also calculated. RESULTS: The DenseNet algorithm model yielded an AUC of 0.99 (95% CI: 0.958-1.0) in the validation set and 0.98 (95% CI: 0.972-0.995) in the test set. The threshold value was selected as 0.8, while for validation and test sets, the accuracies were 95% and 92%, the sensitivities were 100% and 97%, the specificities were 90% and 87%, and the F1 values were 95% and 93%, respectively. The sensitivity of radiologist was 94%, the specificity was 96%, while the accuracy was 95%. CONCLUSIONS: Deep learning (DL) with DenseNet can accurately classify COVID-19 on HRCT with an AUC of 0.98, which can reduce the miss diagnosis rate (combined with radiologists' evaluation) and radiologists' workload.